Listening habits on iPod

[jon abbey]

...but use lossless AIFFs for anything in the electroacoustic realm...

Do you know if there's much of a noticeable difference between AIFF and Apple Lossless? I was using AIFF for all my eai stuff, but when my ipod got full (80 gig), I started removing some stuff and and ripping it again with Apple Lossless. I can't tell much of a difference, but I'm curious if others can.

I had this argument with Robert when I first got an iPod a couple of years ago. I used Apple Lossless for a bit but could definitely tell a difference. no matter how you slice it, they're discarding half of the information, and that's unlikely to all be unnecessary bits in electroacoustic music. Robert says that algorithms say they're identical, but my mastering engineer Earl Howard agrees with me, he says there's a definite and obvious difference.

[Vincent Kargatis]

[no matter how you slice it, they're discarding half of the information

I don't know anything about Apple Lossless, but what is this statement based on? You can't even start to claim "lossless" if you discard any information.

[faster]

Apple claims that audio files compressed with its lossless codec will use up "about half the storage space" that the uncompressed data would require. Testers using a selection of music have found that compressed files are about 40% to 60% the size of the originals depending on the kind of music, similar to other lossless formats. Furthermore, the speed at which it can be decoded makes it useful for a limited-power device such as the iPod.

http://en.wikipedia.org/wiki/Apple_Lossless" onclick= "window.open(this.href);return false;" onclick= "window.open(this.href);return false;

[Salty Swift]

...but use lossless AIFFs for anything in the electroacoustic realm...

Do you know if there's much of a noticeable difference between AIFF and Apple Lossless? I was using AIFF for all my eai stuff, but when my ipod got full (80 gig), I started removing some stuff and and ripping it again with Apple Lossless. I can't tell much of a difference, but I'm curious if others can.

I had this argument with Robert when I first got an iPod a couple of years ago. I used Apple Lossless for a bit but could definitely tell a difference. no matter how you slice it, they're discarding half of the information, and that's unlikely to all be unnecessary bits in electroacoustic music. Robert says that algorithms say they're identical, but my mastering engineer Earl Howard agrees with me, he says there's a definite and obvious difference.

Few questions beg to be asked.
Does Earl listen to music that he masters on an iPod?
Does the Lossless format eat up that much space as opposed to other MP3 formats?
If so, how much more and is the space really worth the trade-off for the sake of noticable quality?

[Vincent Kargatis]

http://en.wikipedia.org/wiki/Apple_Lossless"

If Jon was referring to that statement, he misspoke. Storage space is not information. If there's a difference to be heard (I don't know, though I'd probably only trust blind test results), then it's presumably in the decoding algorithm, and not in a fraudulent claim of "lossless" (at least, that strikes me as less likely).

[jon abbey]

http://en.wikipedia.org/wiki/Apple_Lossless"

If Jon was referring to that statement, he misspoke. Storage space is not information. If there's a difference to be heard (I don't know, though I'd probably only trust blind test results), then it's presumably in the decoding algorithm, and not in a fraudulent claim of "lossless" (at least, that strikes me as less likely).

yeah, I was just referring to actual file size. how is storage space not information? my understanding was that any compression algorithm works by discarding what the software thinks is repetitive elements, and with an Apple Lossless file being half the size of a straight CD rip to AIFF, it seems clear to me that something more than empty storage space is being removed.

anyway, I did enough experimenting with it to know that I wasn't comfortable with using it as a CD equivalent, especially for material I needed to listen to for Erst-related reasons.

[Damon_Smith]

We had this discussion at Bagatellen. I do agree with Jon that mixing and mastering decisions should NOT be made with these formats. I might put one of my recordings on the ipod to decide about which tracks to use, based on form and content, but that is it.
As far as whether or not to release and distribute music in mp3 or other compressed formats, as label owners it just isn't our decision to make, at this point it is just like busking and our only decision is whether to put a hat out or not in other words any music worth listening to is going to end up on mp3, we can choose to make it possible to get some money for them or not.

[Richard Pinnell]

I have listened very very carefully in the past and can tell absolutely no difference at all between AIFF and Apple Lossless formats. The difference between a high-end Mp3 and Apple Lossless is very obvious to me, but I can honestly say that CDs I have burnt from Apple Lossless files sound identical to me to the original CD. If there is a difference between AL and AIFF then it is so negligible that I can't hear it.

[jon abbey]

Earl H explained to me the technical reasons why AL sounds different from AIFFs the last time we went through this, but I don't really remember, I'll ask him again next time I talk to him.

[SOZ]

My limited understanding was that the additional information stored in AIFF was used for music analysis and editing software, but was not used for sound reproduction. But I'll obviously defer to the pros on the matter. My decision was subjectively based on the fact that I personally couldn't tell the difference. I only compared a couple of albums though. Doris was one of them.

Fwiw, I definitely do hear a difference between AIFF or AL and high-end MP3 even with pop music. I unintentionally ripped a Manu Dibango cd to AIFF and it sounded incredible.

Also, I don't know if this makes a difference but I use those fancy schmancy Bose in-ear headphones (earbuds)

[Joda]

I don't really know much about apple's lossless format, but I would assume that there are imperceivable gaps in the information, that our ears fill in, much like sample rates. Which is to say that whether or not you can hear the difference, it's still missing information. It's the same a digital photo, when you zoom in you can see the pixels, but at a distance the eye makes the picture whole. The higher the resolution, the more accurate the representation. Especially if you like music with high frequencies, then it should be important to behold the highest quality possible in a listening experience, representing the frequencies that we can't actually hear, as they inform the other ones that we can hear.

[joe grimm]

there is some misinformation here. lossless is lossless. every one and zero is retained -- it's just packed differently for storage, and then decoded back to full resolution during playback. if you can hear a difference, that difference is coming from some other point in the signal chain (perhaps the d/a converters at your ipod's output jack).

[jon abbey]

there is some misinformation here. lossless is lossless. every one and zero is retained -- it's just packed differently for storage, and then decoded back to full resolution during playback.

well, I just called my mastering engineer again for some clarification here. according to him, the misinformation is in your post, as Apple Lossless uses a compression algorithm, which isn't always perfect upon decompression, no matter what the tests say. he says that not only can he hear a decided difference at times with Apple Lossless or PC lossless as opposed to a AIFF/WAV or CD file, but that none of his audio professional friends will work with Apple lossless files. he does say that if you burn them to a CD, for some reason that helps and they sound better. FWIW, he doesn't know how FLACs fit into this, he had nothing negative to report about them.

anyway, I don't need to argue about this, I'm personally quite happy with either using MP3s or AIFFs, depending on the situation. I'm just sharing my perspective, and I'm well aware it's not the conventional wisdom.

[joseghast]

I haven't used Apple Lossless, but I do use FLAC both downloading from Dime a Dozen and to send some recordings I've done and I haven't noticed differences. I haven't done an exhaustive comparison though.

As I understand, FLAC is a compression algorithm similar to ZIP but specially tuned for audio so it can reduce a file size more than ZIP without losing quality.

FLAC is specifically designed for efficient packing of audio data, unlike general lossless algorithms such as ZIP and gzip. While ZIP may compress a CD-quality audio file by 10 - 20%, FLAC achieves compression rates of 30 - 50% for most music, with significantly greater compression for voice recordings.

And according to the FLAC website (http://flac.sourceforge.net)

Lossless: The encoding of audio (PCM) data incurs no loss of information, and the decoded audio is bit-for-bit identical to what went into the encoder. Each frame contains a 16-bit CRC of the frame data for detecting transmission errors. The integrity of the audio data is further insured by storing an MD5 signature of the original unencoded audio data in the file header, which can be compared against later during decoding or testing.

Anyway, the proof's is in the listening and you have spirited me to do some tests.

[mudd]

i would certainly worry more about playback devices than compression algorithms, especially considering mr. howard's anecdote about re-burning to cd. while all the information of an aiff should be present in any lossless representation, the playback device at hand might lose some information while doing a real-time playback of the compressed data.

re-converting the files to aiff and for example burning them to cd would certainly avoid this problem.

m

[Richard Pinnell]

i would certainly worry more about playback devices than compression algorithms, especially considering mr. howard's anecdote about re-burning to cd. while all the information of an aiff should be present in any lossless representation, the playback device at hand might lose some information while doing a real-time playback of the compressed data.

Yes this makes perfect sense to me.

[Vincent Kargatis]

yeah, I was just referring to actual file size. how is storage space not information? my understanding was that any compression algorithm works by discarding what the software thinks is repetitive elements, and with an Apple Lossless file being half the size of a straight CD rip to AIFF, it seems clear to me that something more than empty storage space is being removed.

Regardless of particulars about Apple Lossless (about which I make no claim, except that bald lying seems improbable), your logic above does not hold. Lossless compression, generally speaking, simply reduces data redundancy. Here's a trivial example:

"1111111111" can be losslessly compressed "10x1" - there, I just reduced the storage size to 40% of the original. There is no loss of information, but a smaller storage space. Compression algorithms needn't throw away data, they can just encode it to take up less space. Throwing away data obviously can increase your space savings.

[jon abbey]

i would certainly worry more about playback devices than compression algorithms, especially considering mr. howard's anecdote about re-burning to cd. while all the information of an aiff should be present in any lossless representation, the playback device at hand might lose some information while doing a real-time playback of the compressed data.

re-converting the files to aiff and for example burning them to cd would certainly avoid this problem.

yeah, this makes sense to me also. my issue is with straight playback of Apple Lossless files in iTunes, and the problem there certainly may be primarily or totally with iTunes. but I'm not trying to find any definitive truths, just a system that works for me, and I stick to MP3s and AIFFs in terms of digital files (if I DL FLACs, I usually convert them to AIFFs before playing them).

[Vincent Kargatis]

Regarding computer playback, the hardware can make a big difference too. A decent soundcard (that measures well) will cost hundreds of dollars, afaik, and most people don't have them.

Arny Krueger is fairly renowned in working this area.

[Joda]

Regarding computer playback, the hardware can make a big difference too. A decent soundcard (that measures well) will cost hundreds of dollars, afaik, and most people don't have them.

Nor do most people have rooms that are acoustically designed to correspond to $10 000+ speakers, if you want to get into specifics about audiophile playback. A home comparison is going to be inevitably coloured by many variables unless you know what to listen for, with the right (or at least expensive ) chain of equipment and room.